A paper published in Nature last November [1] provoked a furore of responses from the pro-biotech community. The journal succumbed to pressure by issuing a retraction: "In light of ..discussions and the diverse advice received, Nature has concluded that the evidence available is not sufficient to justify the publication of the original paper." But, as the authors wish to stand by the evidence and conclusions, Nature thought it best to publish the criticisms, the authors' response and new data, and to let readers "judge the science for themselves." [2]

The criticisms appear to hinge on the experimental techniques used by Berkeley scientists David Quist and Ignacio Chapela to support their claim that transgenic DNA has polluted the Mexican landraces. First, polymerase chain reaction (PCR) enabled them to identify the cauliflower mosaic virus (CaMV) 35S promoter in the landraces. This piece of DNA is incorporated in virtually every commercial transgenic crop. Then, inverse PCR (iPCR) was used to look for unknown DNA sequences joined to the CaMV promoter, which would give information on the structure of the transgenic DNA and its precise location in the genome. PCR is a standard technique, widely employed for amplifying and identifying specific sequences present in trace amounts. Inverse PCR, on the other hand, is a much newer technique, and not yet widely used.

The critics do not take issue with the identification of the CaMV 35S promoter, thereby implicitly acknowledging the presence of transgenic DNA in the landraces. In other words, they are not disputing that transgenic pollution has occurred.

Rather, their criticisms center on the iPCR technique for identifying unknown DNA sequences linked to the CaMV promoter, which they regard as "suspect" and "artifactual".

Quist and Chapela have found a diversity of sequences linked to the promoter, thus giving the impression that the transgenic constructs were "fragmenting and promiscuously scattering throughout genomes", which "would be unprecedented", according to the first critique [3]. It also denies that transgene fragments can move around the genome after integration, and does not bother to tell us that there have been no experiments done previously to address the issue.

The first critique comes from microbiologist Mathew Metz, former colleague of Quist and Chapela, now in University of Washington, Seattle, and Johannes Futterer, from Institute of Plant Science, ETH, Switzerland. The second critique comes from six colleagues of the authors in Berkeley. Berkeley's bioscience department was taken over by biotech giant Novartis in a controversial bid a few years ago, and Ignacio Chapela attracted attention as a major opponent of the take-over. There is no doubt that the attack on Chapela is at least partly motivated by politics, a charge levelled against Chapela's work by his critics from Berkeley. But fortunately, politics is irrelevant in considering what the experimental results are telling us.

PCR and iPCR both depend on short stretches of DNA, called primers, which pair up (or hybridise) with parts of the longer sequence to be amplified. This then enables the DNA copying enzyme to make the rest of the sequence. Unfortunately, the primers often have sequence similarity to other DNA, and so they could hybridise to the wrong places, leading to wrong sequences in the plant genome being amplified. The primers used do have similarities (homologies) to known plant gene sequences, and hence false priming and misidentification of sequences could have given the impression that the CaMV 35S promoter is scattered throughout the genome.

Inverse PCR "is prone to generating artefacts" [4], the second critique states, and a more standard technique, such as "Southern blots of individual kernels" should have been used. Southern blot depends on cutting DNA into pieces with different DNA-cutting enzymes (restriction enzymes), separating the pieces by running them through a gel under the influence of an electric field (electrophoresis), and probing with transgenic DNA to identify the pieces (bands) carrying the transgenic DNA.

In their reply, Quist and Chapela acknowledge that some, though not all of the iPCR results could represent false priming and misidentified sequences [5], and point out that such problems are inherent to the technique. However, that does not alter their  main conclusions.  They provide new data based on a dot-blot technique. A measured amount of DNA is transferred to a filter (in a dot), dried, and then probed with transgenic DNA; in this case, the CaMV 35S promoter.

The new data clearly show the presence of CaMV 35S promoter in four landrace samples at levels less than 5% and greater than 1%, while a historic maize sample and a maize sample from Peru both stain negative. In other words, transgenic pollution has  indeed occurred as reported in their previous paper.

The real disagreement is to what extent the transgenic constructs had fragmented on entering the genome of the landraces, or thereafter. The existing evidence on transgenic instability, documented in some papers cited by Quist and Chapela [5], does not rule out the possibility of "fragmenting and promiscuous scattering" of transgenic constructs, which could have introgressed into landraces via horizontal gene transfer as well as by cross-pollination [6]. The significance of Quist and Chapela's work [1] is that it is the first of its kind in attempting to address this possibility.

Once again, the scientific establishment serving the corporate agenda has been caught out taking the absence of evidence as evidence of absence. The agenda is to keep the public confused while transgenic pollution continues unabated. 

Above all, corporate scientists want to avoid having to prove transgenic lines are stable by the appropriate 'event-specific' molecular data that the new European Directive requires [7]. This involves documenting that the transgenic insert has maintained the same structure and location in the plant genome in successive generations. No such 'event-specific' molecular analysis has ever been done for any transgenic line. Significantly, Monsanto's Roundup Ready GM soya failed the test when recently analysed [8]. Regulators should insist on this molecular data, and the data should not be hidden away from the public under "commercial confidentiality". Otherwise, regulators should be held liable for any damages caused as a result.

The only decent thing for the scientific establishment to do now is to give plenty of support to Quist and Chapela and others to extend their research. The aim is to rule out the possibility that transgenic constructs could be fragmenting and scattering, throughout the genome as well as throughout the ecosystem, by horizontal gene transfer and recombination. Meanwhile, no more transgenic crops should be released, especially those with the CaMV 35S promoter [9], until they could be proved stable by event-specific analyses.

1.Quist, D. & Chapela, I.H. Transgenic DNA introgressed into traditional maize landraces in Oaxaca, Mexico.    Nature 2001,  414, 541-3.
2.Editorial note. Nature Advance online publication Nature , 4 April, www.nature.com
3.Metz M and Futterer J. Suspect evidence of transgenic contamination. Nature Advance online publication Nature , 4 April,  www.nature.com
4.Kaplinsky N, Braun D, Lisch D, Hay A, Hake S, Freeling M. Maize transgene results in mexico are artefacts.
5.Quist and Chapela reply. Nature Advance online publication Nature , 4 April, www.nature.com
6.See Transgenic Instability, ISIS reprint collection, ISIS Members website                       http://www.i-sis.org.uk/full/pdf/ISIS_REPRINTS_transgenic_instability.pdf
7.See "Europe's new rules could sink all GMOs" by Angela Ryan and Mae-Wan Ho, ISIS News 11/12, October 2001
http://www.i-sis.org.uk/isisnews/i-sisnews11-3.php
8.Windels P, Taverniers I, Depicker A, Van Bockstaele E and De Loose M. Characterisation of the Roundup Ready soybean insert. Eur Food Res Technol DOI 10.1007/ s002170100336, © Springer-Verlag, 2001; see also "Scrambled genome of Roundup Ready soya" by Mae-Wan Ho, Transgenic Instability, ISIS Reprints, ISIS Publications, London; also on ISIS members website.
9.Ho MW, Ryan A. & Cummins J. The cauliflower mosaic virus promoter - a recipe for disaster? Microbial Ecology in Health and Disease 1999, 11, 194-197 http://www.i-sis.org.uk/camvrecdis.php; Ho MW, Ryan A. & Cummins J. Hazards of transgenic plants with the cauliflower mosaic virus promoter. Microbial Ecology in Health and Disease 2000, 12, 6-11 http://www.i-sis.org.uk/camv-mehd.php; Ho MW, Ryan A. & Cummins J. CaMV 35S fragmentation hotspot confirmed, and  it is active in animals. Microbial Ecology in Health and Disease 2000, 12,189 http://www.i-sis.org.uk/mehd3.php; see also Ho MW & Cummins, J. Who's afraid of horizontal gene transfer? ISIS Report, 4 March, 2002 http://www.i-sis.org.uk/hgt.php.